The goal is not extreme life span but something they call “health span.” Rather than getting heart disease or cancer in our fifties or sixties and needing expensive treatments and drugs to keep us alive (if they do) until we’re 75, we will age more slowly. We will still get these chronic diseases, but 10 or 20 years later, shortening considerably the time we spend in hospitals and nursing homes and the money we, and society as a whole, have to spend on health care. “What we would like to do,” Sharp says, “is prevent a lot of things that would put you into the hospital prematurely. The end is likely to be bad no matter what, but we’d like to compress the time that we have to do this intensive type of stuff.”

That both health span and life span are to a large extent determined by our genes is one of the more obvious and yet unexpected findings to come out of recent studies of centenarians, the oldest of the old. It is obvious because different species of animals age at different rates. Mice will last about three years, but after two they are likely to be getting frail or starting to succumb to cancer, organ failure, and diabetes. The same things happen in dogs at around 10 years and in horses at closer to 20. Chimpanzees take 30 years and humans typically 60 or more before the process of decrepitude begins. Rats and tree squirrels are virtually the same size and often live in the same environments, yet rats will live 3 or 4 years and squirrels as many as 20. Clearly some genetic mechanism must be determining how quickly each animal ages.

Still, when Barzilai and his colleagues questioned the near centenarians as part of their aging studies, the genetic determinant of longevity took the researchers by surprise. “When we started recruiting 100-year-olds,” Barzilai says, “we noticed something interesting. They have a family history of longevity. When we ask them, among other things, ‘Why do you think you lived to be so old?’ they usually say, ‘What do you mean? All my brothers and sisters are over 100.’ Or ‘My mother was 102; my grandfather was 108.’

“Then we say, ‘OK, tell us really the truth. You ate yogurt your whole life. You were a vegetarian.’ But the interesting thing is, we have only 2 percent vegetarians. We have none who exercised regularly, and 30 percent were overweight or obese back in the 1950s, when not that many people were overweight or obese. Almost 30 percent have smoked two packs of cigarettes for more than 40 years. We have one woman, alive now—she’s 107—who celebrated 95 years of two-pack-a-day cigarette smoking. She had four siblings in my study. All of them were over 100. One younger sister died at 102, the poor thing.”

This is not to say that smoking cigarettes won’t prematurely kill the rest of us, or that exercising regularly won’t make us live longer. “What I tell people,” says Steven Austad, an expert on the biology of aging at the University of Texas Health Science Center, “is that if you want to live to be a healthy 80-year-old, you have to eat right and exercise, et cetera. If you want to live to be a healthy 100-year-old, you have to have the right parents.” The ultimate aging clock moves by order of our genes. The genes can be influenced by the environment, lifestyle, and diet, but only finding the genes and deciphering their action will allow us to bottle and distribute the sauce.

Despite the recent explosion of life-extension research, the single most important observation in the field—the best clue about the nature of that special sauce—was first made three-quarters of a century ago. In 1935, the Cornell University nutritionist Clive McCay reported that feeding rats just barely enough to keep them alive extended their life span by as much as 50 percent. Several years later, a Chicago pathologist named Albert Tannenbaum found the same thing in mice. Feed these animals little more than a starvation diet, he noted, and it dramatically inhibits tumors. In one experiment, 26 of 50 well-fed mice developed mammary tumors after two years, compared with zero of 50 that were allowed only enough food to keep them going. Tannenbaum’s semi-starved animals not only lived longer but were more active, he reported, and had fewer “pathologic changes in the heart, kidneys, liver, and other organs.”

Most researchers paid scant attention to McCay’s and Tannenbaum’s results until the early 1990s, when the underlying genes were found. That’s when longevity research took off. The effect of calorie restriction on health and longevity has been shown to hold true not just for rodents but also for yeast, protozoa, fruit flies, worms, spiders, and maybe monkeys. The intervention prevents heart disease, cancer, diabetes, kidney disease, cataracts, Parkinson’s, and Alzheimer’s. It improves cholesterol profiles, lowers blood pressure, and prevents the deterioration of the immune system that naturally accompanies aging.

Calorie restriction may also preserve intellectual function, or at least what passes for intellectual function in laboratory rodents. The maximum life span of a normally fed lab mouse is a little more than three years. Semi-starved lab mice will not only live about 25 percent longer but will run a maze at three years with the facility of a well-fed mouse of six months.

Researchers cannot tell whether calorie restriction extends human life, because our species tends to live too long to make this kind of experimentation possible. But they are trying to see if it improves at least the biomarkers of health status—blood pressure, cholesterol profiles, and the like. Preliminary reports indicate that a reduced diet does produce such improvements, at least over the short term of the studies. But it is hardly a viable strategy for improving public health for an obvious reason: Precious few of us have the willpower—or desire, for that matter—to voluntarily subject ourselves to this kind of lifelong regimen. And clinicians working with anorexics have reported that after a decade or two, those with partial or subthreshold forms of the disorder—who maintain a diet some have compared to calorie restriction—show a failure to thrive, with damage to hearts, lungs and other organs.

Ultimately, antiaging researchers used calorie restriction not as a formulation for a longevity diet but as a lens through which we might study and mimic the mechanisms of extended youth. The hypothesis that calorie restriction works because semi-starved animals have less body fat and so do not have to work as hard to stay alive, fell out of favor in the 1980s. Back then, Harrison restricted calories for genetically modified obese mice, which remained extremely fat even when food was cut. These mice lived just as long as semi-starved lean mice, even though they had at least four times the body fat of normal mice. This experiment suggested that longevity in calorie-restricted animals was “related to food consumption rather than to the degree of adiposity,” Harrison wrote at the time.

Ever since, longevity researchers have explored a lot of ideas about what eating less actually does. One hypothesis proposes that calorie restriction reduces the creation of toxic molecules called free radicals, which are considered crucial factors in the aging of cells and tissues. (Free radicals oxidize other biological molecules, the same thing that happens to a metal when it rusts.) If an animal eats less food, the concept goes, its cells burn less fuel—glucose, in particular—and generate fewer free radicals. As a result, oxidative stress proceeds at a slower pace and the organism lives longer, just as a car will last longer in a dry climate that doesn’t promote rust. One thing researchers agree on is that calorie restriction reduces the number of free radicals floating around, and so presumably the amount of oxidative stress. And it seems clear that when fruit flies are either fed antioxidants or genetically manipulated to overproduce their own, they live up to 50 percent longer than normal.

So far, so good. But boosting the antioxidant supply in rodents benefits them not at all. Nor have clinical trials supported the popular belief that humans benefit from taking antioxidant supplements or eating antioxidant-rich foods (despite the breathless articles and infomercials pushing the benefits of everything from chocolate, green tea, and açai to more mundane fruits, vegetables, and legumes). These results began to turn researchers away from the free radical theory of aging.

The animals known as naked mole rats may have turned them away for good. These hairless rodents come from Kenya, where they live in underground colonies that can be miles across and contain as many as 300 animals. Rochelle Buffenstein, a physiologist who is now at the University of Texas Health Science Center, began studying naked mole rats in 1980 and took a colony with her when she moved from South Africa to the United States in 1997. Nearly two decades after she began, many of her original animals were still going strong. “Most animals that size live no longer than three years,” she says. “I started thinking, my gosh, some of these animals are now 17 years of age. And so I started doing research on aging.” Buffenstein and others have established that naked mole rats will live as long as 30 years, staying perfectly healthy for most of that time. “They resist the whole aging phenomenon,” she says, until they’re 24 or 25 years old, the equivalent of 102 human years. Until then, Buffenstein notes, “you can’t tell a young guy from an old guy.” Only after 26 years do they start losing muscle mass and showing signs of decrepitude and imminent demise.

However it is that naked mole rats manage to stay young and vigorous for so long, minimizing oxidative stress has little or nothing to do with it. “Even very young animals have very high levels of oxidative damage,” Buffenstein says. “In fact, they exceed that of mice of equivalent age, although mice live another two years and these guys another 26 to 28. That throws a spanner in the works of the oxidative stress theory of aging.”

Another hypothesis that has received an enormous amount of media attention but that now seems to be losing credibility is that the benefits of calorie restriction are mediated through a set of purported antiaging genes called sirtuins. These genes were first linked to aging by MIT biologist Leonard Guarente as bestowing an unusually long life span in yeast. Guarente’s former postdoc David Sinclair, now at Harvard, reported in 2003 that resveratrol, a compound found in red wine, is capable of “hyperactivating” the sirtuin gene. Some nutritionists had already speculated that red wine might be the explanation for what is known as the French paradox—that the French are particularly long-lived and relatively free of heart disease despite their fat-rich diets. So it made sense when it turned out that resveratrol might be a potent antiaging drug and that sirtuins might be key players regulating the benefits of calorie restriction.

By 2004 Sinclair and his business colleagues had established a biotech company called Sirtris to identify drugs even more effective and long lasting than the reversatrol-sirtuin interaction. Two years ago the pharmaceutical giant GlaxoSmithKline purchased Sirtris for $720 million, and by August of last year The New York Times was speculating that compounds capable of activating sirtuin genes were now the leading candidates for what the newspaper called “the ultimate free lunch…a drug that tricks the body into thinking it is on [a calorie-restricted] diet.”

But the latest research suggests that the resveratrol-sirtuin story is significantly less promising than has been portrayed, or perhaps even dead wrong. In the past year, research teams from two pharmaceutical companies, Amgen and Pfizer, published studies indicating that much of the original work on resveratrol and sirtuin genes was misinterpreted. The Pfizer study, says Gary Ruvkun, a geneticist and molecular biologist at Harvard Medical School and Massachusetts General Hospital, “basically says we don’t see resveratrol doing anything. When we treat mice with resveratrol, it doesn’t do anything other than make them sick.” Ruvkun described the result as “part of the crumbling empire of sirtuins,” an opinion shared widely in the field.

The one explanation for how calorie restriction works that has remained intact is that it decreases the secretion of certain hormones—in particular, insulin and insulin-like growth factor—that signal organisms to channel their resources into either growth and reproduction (when insulin and IGF levels are high) or maintenance and repair (when they are low). In the vernacular of the science, calorie restriction appears to increase life span and health span at least in part by “reducing signaling in the insulin/IGF pathway.”

This line of thinking is built on a series of breakthroughs that began in the early 1990s. At the University of California, San Francisco, geneticist Cynthia Kenyon and her colleagues discovered a genetic mutation in nematode worms that doubles their life span. At the time, this was the longest extension of life span ever observed. In 1997 researchers in Ruvkun’s laboratory at Harvard Medical School reported that the gene in question was the worm equivalent of a trio of insulin-related genes in humans. By 2003 other scientists had genetically manipulated mice to be relatively insensitive to either insulin or insulin-like growth factor; in both cases, the genetically engineered mice lived significantly longer than normal mice.

Insulin and insulin-like growth factor happen to be two hormones that respond to the presence or absence of food in the environment. Put simply, they respond to how much we eat: The more we eat, and particularly the more sugar and starch we eat, the more insulin the body secretes in response. Together, these two hormones regulate metabolism, fat storage, and reproduction. Insulin-like growth factor promotes cell division and growth, while insulin shunts fuel consumed either into immediate energy use or into storage for a later time. When food is plentiful, insulin and IGF levels go up and signal the animal that it is OK to grow, mature, and reproduce. When food is scarce, insulin and IGF levels go down. Activity in the insulin/IGF signaling pathway is reduced, and the animal shifts into a maintenance mode that favors long-term survival over immediate reproduction. The outcome is a redirection of resources toward repairing and protecting cells.

“When food becomes limiting, an animal lacking this system would either die of starvation or produce progeny that die of starvation,” Kenyon says. “In contrast, with this food-sensing system in place, the animal builds up reserves and suspends reproduction until food is restored. It also activates pathways that extend life span, which increases the organism’s chance of being alive and still youthful enough to reproduce if it takes a long time for conditions to improve.”

It is with insulin and IGF that the genetics and biology of aging are finally coming together. The most obvious genetic difference between long-lived and short-lived strains of mice is variation in the insulin/IGF gene. The longest-lived strain of nematode worm on record, called AGE-1, can live 10 times longer than normal worms. And the mutation that bestows such remarkable longevity is a gene that happens to regulate activity in the insulin/IGF signaling pathway, says Robert Schmookler-Reis of the University of Arkansas, a molecular geneticist who discovered the relationship between the mutation and the superworm’s extraordinary life. That mutation helps determine how the various cells in the worm respond to insulin and IGF signaling. Even rapamycin regulates way cells respond to insulin and IGF, albeit in this case by turning up the pathway rather than turning it down.

The genetic markers for life span in any given person, however, are not so clear-cut. When researchers like Barzilai have looked for genes that might account for the extreme longevity of the centenarians they study, they have typically found that the genes that stand out in one long-lived population do not do so in others. This is not so surprising, explains Austad, because genes don’t operate in isolation; they operate in response to their environment. And when researchers look at different populations of individuals who have lived past 90 or 100, they are looking at people who managed to survive vastly different environmental stresses. “Some went through World War II and had to survive famine in Europe,” Austad says. “Some had to first survive the 1918 Spanish flu. And so some genes may benefit survival only under certain circumstances.”

The most “striking” find, Austad adds, is the role of a gene known as FOXO, which is regulated by the insulin/IGF pathway. A variant form of the gene was first noted in humans in a Hawaiian study of long-lived men of Japanese ancestry and has since been found in long-lived Germans, Italians, Ashkenazi Jews, and Chinese. “This is an amazing finding,” Kenyon says. “We don’t know how much we can change our life span by fooling around with [this FOXO gene], but at least we know that humans are susceptible to the effect of this pathway.”

THE END OF AGING?

If longevity researcher Aubrey de Grey has his way, the life span of a healthy human would hit 1,000 years. Chief science officer at the SENS Foundation (Strategies for Engineered Negligible Senescence), the wildly bearded de Grey says that other scientists are thinking too small because they have become mired in reductionist mechanisms and secondary details about how aging works. Rather than tinker with the mechanics of some insulin pathway or hundreds of obscure genes, he wants to clean up all the molecular garbage that is created by normal metabolic processes (like breathing and eating) but not excreted or destroyed. As a person ages, the accumulated junk—including cholesterol-related molecules in blood cells and proteins aggregated in the brain—increasingly interferes with health, de Grey argues, “just as your house won’t work well if you don’t take out the trash for a month.” Cardiovascular disease and neurodegenerative diseases like Alzheimer’s and Parkinson’s result from the buildup, in this view.

De Grey’s battle plan? In a one-two punch, he would deploy microorganisms to dismantle and haul out the molecular trash while delivering engineered gene and therapeutic cells to refurbish cells that have died out and gone unreplaced. His final step would be to combine all the techniques into one injectable brew. When such a treatment is available, “adults will die only of the same causes that younger people die from today,” he says.

Gary Taubes writes for Science, The New York Times Magazine, and many other publications. He was a staff writer for DISCOVER in the 1980s and is the author of the best seller Good Calories, Bad Calories.